What did you do with the data? Please describe every data reduction or data analysis exercise step by step.

DAY 1: We divided into two teams. Lauren, Howard and his students used Leopard to download a spectral file and to process it into line spectra. My students and I used Leopard to download the peakup images. We then used IDL to align and stack them. Just before lunch, we had planned to exchange procedures with the other team. However, we ran out of time and only got the briefest of explanations about what the other team had done. During the IDL data reduction, our students did most of the keyboard work. They seemed to catch on to the nuances of IDL faster than the teachers did. It would have been helpful to have a list of commonly used IDL functions.

Following lunch, we went on a walking tour of the CIT campus. Upon returning from the tour, we had a meeting with the Spitzer Center Director.

After the meeting, we continued with the data reduction process. With Ranga's help, my group performed photometry on the stacked image, obtaining a brightness count and a sky reading for the galaxy nucleus. Late that afternoon, I talked to Ranga, and asked if it would be possible to move on to data analysis, rather than work on image reduction anymore. I reasoned that the image reduction procedures could not be done in the classroom by us or by other teachers because they required UNIX operating systems, IDL software, and a degree of proficiency with IDL. Ranga agreed to begin the process of reducing the spectral .FITS files to line spectra.

DAY 2: When we arrived on day 2, Ranga had left us a homework assignment. He asked us to 1) determine the diameter of the galaxy nucleus, 2) determine whether a black body curve could be fit to three known data points (from the Keck telescope, we knew that ARP102B had an intensity of 90 mJy at 12 microns, but dropped below the detectable threshold of 50 mJy at 17.5 microns. From our photometry of the peakup image, we knew that the intensity at 16 microns was 54.5 mJy. ) 3) calculate the spectral index of the energy source using the three known points.

Ranga had already applied the platescale to the peakup image and converted the size of the nucleus to an angular width in arcseconds. I showed my students how to use the small angle approximation to calculate the diameter of the galaxy nucleus. They got the same answer I got.

I then built a blackbody model in Excel and determined that a blackbody at 345K could fit the data points. Later when I showed the model to Ranga, he suggested that the continuum in our spectrum was probably a composite of several black bodies rather than a single one.

I did not know how to calculate a spectral index, but when Ranga came in he showed me how to do it. He also pointed out a mistake I had made in my blackbody calculation, and I spent the rest of the morning correcting it. Ranga gave Howard and Lauren a problem to work involving the aperture of the telescope. I am not sure what the problem was, but I believe they did not get it finished. We wondered if he gave it to keep us busy while he worked on the spectral reduction.

After the diameter calculation, we did not have anything the students could do, so they read, played video games, and talked. Eventually, Doris came in and took them to see how pseudo-color images and posters were made using PhotoShop. Late that afternoon, Ranga finished processing the first spectral data, the SL mode data. I quickly imported it into Excel and adjusted the lines for Doppler shift. Lauren, Howard, and I then began to work out procedures that the students could use to identify and extract line data. I set the students to search the Internet for a source for spectral line identification.

That night after I returned to the hotel from dinner, I found a spectral line table in the Handbook of Infrared Astronomy and loaded it into an Exel lookup table.

DAY 3:

When we arrived on Day 3, Ranga had completed the processing of a second spectral file, the SH mode spectra. We transferred it from the FTP, imported it into Excel, and set the students to working on extracting line ID's, expected and actual wavelengths, peak flux, and FWHM. Howard's students worked on the SL data. My students worked on the SH. Lauren and Howard began extracting sky data at various wavelengths using SPOT. At about 11 AM the media team arrived from the local newspaper and began interviewing us. When they left, the students went to lunch, and upon their return, went on the Mt Wilson tour with Howard and Doris. At the same time, Ranga advised that he had finished processing the third spectral data file. Lauren and I downloaded it and spent the remainder of the afternoon working together on it. (We were down to one laptop: the AC adapter failed on mine, and neither of us was comfortable working with Howard's Mac).

Late that afternoon, Ranga sat down with Lauren and I and we looked at the spectra. Ranga showed us how to use the lambda/delta lambda calculation to look for Doppler shifted lines. We also discussed in general the analysis procedures we would need to do when we got home:

1) We need to subtract background (sky) from the flux values in our spectra.
2) We need to finish identification of spectral lines. Ranga pointed out that the table we were using was only atomic spectra - we need to find a table for molecular spectra, too.
3) We need to fit Gaussian curves to key diagnostic lines and find their flux from the area under those curves.
4) We need to calculate separation of lines to see if they are Doppler shifted and compare with the hydrogen Balmer line splits previously reported.
5) We need to calculate ratio of fluxes for key diagnostic lines.
6) We need to calculate dust column density from the Silicate absorption line. H to Silicate ratio gives dust column density.
7) Plug data into CLOUDY to get energy sources and geometry of the torus.
8) Eventually, need to link UV, Optical, IR, and X-ray spectra to make a complete SED.
9) Figure out if we have enhanced understanding of SMBH formation and evolution. If so, we could do a science journal article. Might take a couple of years.

After looking at our spectral lines, Ranga decided he needed to do more smoothing to the data files to remove some more of the noise.

What do you plan to do with the data when you return? As part of data reduction or data analysis? Please list all exercises you plan to do.

Actually, Lauren and I want to get as much of the analysis done this summer as possible. We won't have much time to work on the project after school starts. We plan to leave the Doppler shifted lines for our research students to analyze this fall. We want them to write an article for the RBSE Student Research Journal. I've told my students that they really need to finish the article this fall so that they can include its submission as part of their college resumes.

After I returned home, I used Excel to create a Gaussian curve fitting procedure. I tried it out on a couple of lines and emailed them to Ranga. He said the procedure appeared to be okay. Lauren emailed me the background fluxes that he worked on in Pasadena. I'm going to check them out this week using Spot. Hopefully, we will soon have the refined spectral files from Ranga with more of the noise removed. When we get them, I'll subtract the background and we'll start calculating flux for key diagnostic line ratios. Among the first flux ratios we will do is H to Silicate absorption so that we can get the dust column density. Concurrently, we will need to identify some of the unknown spectral lines. By the end of summer, I'd like to have all the line ratios complete.

Next, we will need to locate line spectra from the UV, optical, and x-ray studies. We will then normalize these data before combining them. I'd like to have a multi-wavelength SED by October. In October, we need to have a teleconference with Ranga to assess what we've learned, then write our AAS abstract. In Oct and Nov, we need to put together our AAS poster and handouts.

Over the Christmas - New Years break, I'd like to begin running CLOUDY simulations.